>> What do you want to do with Spectrum programs that test keyboard >> with 0 in high byte? > >Hi byte is 0 mostly for a PRESS 'ANY KEY' test. >I would map row SPACE SS M N B as the first row to scan. >So one of these keys can be pressed in a ANY KEY situation.
Hmmm... the way Martin and I came up for doing this was this: Take a Z80 processor (any speed), and an 8K eprom, plus a few gates to handle I/O with the keyboard. (any I/O port becomes keyboard access). Program the Z80 to read the clocked serial data in (dead easy) from the keyboard, decode the scan codes (scan mode 3 is the best one to use -- it gives single code lookups for every possible key combination, IIRC -- check the Zilog keyboard controller site for more details). Put these scan codes into a 255 byte buffer, doing all the necessary combinatorics. (255 byte, as the last one is not reading any key data, IIRC, so you just float high). Hey presto! A little wait-state logic, and you're sorted. We had a prototype at one point -- worked quite well. I had the keyboard LED's cycling and everything... Simon ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com
